VeenaMAX TTS Provider Integration #1357
Conversation
|
DescriptionSummary By MatterAI
🔄 What ChangedThis pull request integrates the new VeenaMAX Text-to-Speech (TTS) provider into the system. It involves defining a new global constant for VeenaMAX, registering the provider within the existing provider index, and creating dedicated files for its API configuration, parameter mapping, and response transformation logic. The integration includes handling various API responses and errors from the VeenaMAX service. 🔍 Impact of the ChangeThis change expands the platform's capabilities by adding a new TTS provider, offering more options for speech generation. The structured integration ensures maintainability and consistent error handling across providers. It introduces new external API calls to 📁 Total Files Changed
🧪 Test AddedNo explicit test files were added in this PR. The existing test suite should be extended to cover the new VeenaMAX provider integration, specifically for successful speech creation and various error scenarios. 🔒Security VulnerabilitiesThe MotivationIntegration of a new Text-to-Speech provider to expand service offerings and capabilities. Type of Change
How Has This Been Tested?
Screenshots (if applicable)N/A Checklist
Tip Quality Recommendations
Tanka Poem ♫
Sequence DiagramsequenceDiagram
participant Client
participant GatewayAPI as Gateway API
participant VeenaMAX as VeenaMAX TTS Service
Client->>GatewayAPI: POST /v1/speech/create (input: string, voice: string, model: string, ...)
Note over GatewayAPI: Route request to VeenaMAX provider
GatewayAPI->>GatewayAPI: Map parameters using VeenaMaxCreateSpeechConfig
GatewayAPI->>GatewayAPI: Get API endpoint from VeenaMaxAPIConfig.getEndpoint({fn: 'createSpeech'})
GatewayAPI->>GatewayAPI: Prepare headers with Authorization: Bearer {apiKey} from VeenaMaxAPIConfig.headers
GatewayAPI->>VeenaMAX: POST https://flash.mayaresearch.ai/generate (text: string, speaker_id: string, streaming?: boolean, normalize?: boolean)
alt Successful Response (HTTP 200 OK)
VeenaMAX-->>GatewayAPI: Audio Data (e.g., WAV format)
GatewayAPI->>GatewayAPI: Transform response using VeenaMaxCreateSpeechResponseTransform(audioData, 200)
GatewayAPI-->>Client: 200 OK (Audio Data)
else Error Response (HTTP 4xx/5xx)
VeenaMAX-->>GatewayAPI: Error JSON (error: {message, code, type})
GatewayAPI->>GatewayAPI: Transform error using VeenaMaxCreateSpeechResponseTransform(errorJSON, statusCode)
GatewayAPI->>GatewayAPI: Generate standardized error response via generateErrorResponse(...)
GatewayAPI-->>Client: Error Response (e.g., 400 Bad Request, 401 Unauthorized)
end
|
![matter-code-review[bot]](https://avatars.githubusercontent.com/in/474549?s=60&v=4){kind=small}
| code: | ||
| errorResponse.error.code?.toString() || responseStatus.toString(), |
There was a problem hiding this comment.
🐛 Bug Fix
Issue: Error response handling fails when errorResponse.error.code is 0 (falsy), causing undefined in error code
Fix: Use nullish coalescing to handle 0 correctly
Impact: Prevents incorrect error codes in API responses
| code: | |
| errorResponse.error.code?.toString() || responseStatus.toString(), | |
| code: | |
| errorResponse.error.code ?? responseStatus.toString(), |
| ) => { | ||
| if (responseStatus !== 200) { | ||
| if (response && typeof response === 'object' && 'error' in response) { | ||
| const errorResponse = response as VeenaMaxErrorResponse; |
There was a problem hiding this comment.
🛠️ Code Refactor
Issue: Redundant type assertion after in operator check
Fix: Remove unnecessary type assertion since in operator already confirms type
Impact: Cleaner code without runtime impact
| const errorResponse = response as VeenaMaxErrorResponse; | |
| const errorResponse = response; |
|
|
||
| switch (responseStatus) { | ||
| case 400: | ||
| errorMessage = |
There was a problem hiding this comment.
Hey @saicherry93479 I would not recommend keeping a map of the error messages here in the gateway, please forward whatever error message is returned from the upstream provider in the https response
|
Hi @saicherry93479 , would you be able to address the comments? |
3 participants
Description
This PR adds VeenaMAX TTS (Text-to-Speech) provider integration
Type of Change
Changes Made
Core Integration Files
src/providers/veenamax/api.ts - API configuration with base URL, headers, and endpoint mapping
src/providers/veenamax/createSpeech.ts - TTS implementation with parameter mapping and response transformation
src/providers/veenamax/index.ts - Main provider configuration export
src/globals.ts - Added VEENA_MAX constant
src/providers/index.ts - Registered VeenaMAX provider
How Has This Been Tested?
src/providers/veenamax/api.test.ts - Tests API configuration, headers, and endpoint mapping
src/providers/veenamax/createSpeech.test.ts - Tests TTS functionality, error handling, and voice mappings
Basic TTS request
curl --location 'http://localhost:8787/v1/audio/speech'
--header 'Content-Type: application/json'
--header 'x-portkey-provider: veenamax'
--header 'Authorization: Bearer your_veenamax_api_key'
--data '{
"input": "Hello! This is a REST API test of VeenaMAX TTS through Portkey Gateway.",
"voice": "charu_soft",
"response_format": "wav"
}'